The Agent Skills Directory

[SAFE]: The skill's primary functionality is to read local project documents and write structured plans to the project directory. It does not perform network operations or access sensitive system files outside its defined scope.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes data from external files (structure, design, research) and incorporates it into its workflow. Ingestion points: Reads files from 'thoughts/shared/structure/', 'thoughts/shared/design/', and 'thoughts/shared/research/'. Boundary markers: Missing explicit delimiters to isolate ingested file content from core instructions. Capability inventory: Reads and writes project files, spawns auxiliary agents (codebase-analyzer), and generates shell commands for inclusion in human-reviewed plans. Sanitization: No sanitization or validation of the ingested file content is described in the skill instructions.

qrspi-create-plan